473 Analysis of structure and mechanical properties of AA 5083 aluminium alloy processed by ECAE

Corresponding author: N. Llorca-sern, e-mail: nullorca@ub.edu

Rev.Adv.Mater.Sci. 10 (2005) 473-478
2005 Advanced Study Center Co. Ltd.
ANALYSIS OF STRUCTURE AND MECHANICAL
PROPERTIES OF AA 5083 ALUMINIUM ALLOY
PROCESSED BY ECAE
N. LIorca-Isern
1
, C. Luis-Prez
2
, P.A. GonzIez
2
, L. Laborde
1
and D. Patio
1
1
CPCM, Dept. Enginyeria Qumica i Metallrgia, Fac. Qumica, Universitat de Barcelona, Mart-Franqus 1-11,
08028 Barcelona, Spain,
2
Dept. ngeniera Mecnica, Energtica y de Materiales, Universidad Pblica de Navarra, Campus de
Arrosadia s/n, 31003 Pamplona, Spain
Receiced: May 05, 2005
Abstract. n order to alight a little more on the understanding of the mechanisms involved in
nanostructuration when severe plastic deformation is carried out by ECAP on ductile alloys,
different routes have been applied on a AA 5083 alloy at 423K. Metallographic characterisation
has been done by means of optical microscopy, scanning and high resolution transmission
electron microscopy to investigate size, morphology, and distribution evolution of the metallic
matrix and the precipitates present in this type of aluminium alloy. Also, some mechanical prop-
erties have been studied using nanoindentation, US measurements and mechanical testing.
From the results it is worth noting that the microstructure shows two observation levels. n one
case, as expected, micrometric level does follow classical statements and thus properties are
related to microstructure evolution, whereas in the nanometric scale these relationships do not
seem to agree.
1. INTRODUCTION
n recent years there has been an increasing inter-
est in the production of ultrafine-grained material
for both commercial as well as for research pur-
poses by applying severe plastic deformation pro-
cesses [1-4]. The equal channel angular pressing
or extrusion (ECAP) has been chosen in the present
work for the non age-hardenable AA5083 Al alloy.
This Al-3%Mg based alloy exhibits interesting me-
chanical and other physical properties, in particu-
lar their remarkable strength and toughness and
their potential for superplasticity at low tempera-
tures [5]. However, enhancement of strength by
conventional thermal treatment in some non heat-
treatable alloys is impossible [6]; thus, the need of
the use of severe plastic deformation methods. The
present study examines the changes of structure
and mechanical properties of ECAP processed
AA5083 by passing up to 11 times through 90 die
by Route A (no rotation about sample axis between
passes) and up to 4 passes through the same 90
die following Route C (90 rotation about sample
axis after each passage through the die). Microstruc-
tures were analysed by OM, FE-SEM, TEM and
mechanical properties were measured using tensile
test, nanoindentation, and US testing.
2. EXPERIMENTAL PROCEDURE
A commercial AA5083 aluminium alloy (Al-4.5%Mg-
1.0%Mn-0.15%Cr) supplied in the as-cast state
474 N. Llorca-Isern, C. Luis-Prez, P.A. Gonzlez, L. Laborde and D. Patio
material was selected for the ECAE process. Be-
fore being processed, the billets were machined to
cylindrical samples of 10 mm in diameter and 100
mm in length. The ECAE processing was carried
out at a press speed of 50 mm min
-1
and at 423K in
air to restraint recrystallization using a circular cross-
section die with an angle = 90 and ~10 that
produces a deformation of ~1.05 on each passage
through the die. One set of rods was subjected to 4
passes of ECAP by route C, at which the sample
after each pass is rotated along a longitudinal axis
by180, thus in our specimens processed to a com-
plete cycle. Another set of rods was extruded by
route A in which the sample is removed from the die
and the extrusion is repeated without any rotation
until 11 passes. The initial microstructure and its
evolution through ECAP were examined using opti-
cal and electron microscopes, S-4100 Hitachi Field
emission scanning electron microscope (FEG-SEM)
operating at 30 kV and CM-30 Philipps high resolu-
tion transmission electron microscope (HRTEM)
operated at 300 kV. The foils for TEM observation
were prepared from the discs at the centre of speci-
men and were thinned by dual ion-beam milling
(Gatan). The density of dislocations (m
-2
) in TEM
imaging was estimated using Ham and Sharp
method [7] consisting of line-dislocations intercept
counting:
= 2N/Lt, (1)
where N is the number of intercept, L is the total
lines length (m) and t is the foil thickness; t is cal-
culated using Hirsch method [8]. Tensile tests at
room temperature, Vickers hardness, and nanoin-
dentation testing were carried out from samples cut
at the centre of the specimen.
3. RESULTS AND DISCUSSION
3.1. Microstructure of AA5083 before
ECAP
The initial microstructure of the as-cast AA5083
consisted of equiaxed grains with an average size
of about 100 m (Figs. 1a and 1b) in which the den-
dritic segregation structure is clearly seen. t s worth
noting that from inspection of the as-received
sample, coarse precipitates of (Fe, Mn)
3
SiAl
12
and
small Mg
2
Si particles are present in the alloy as
second-phase. Longitudinal section micrograph is
shown in Fig. 1c where similar features to trans-
verse section can be observed. TEM analysis of the
undeformed state (Figs. 1d and 1e) showed high
Fig. 1. nitial microstructure of AA5083 as-cast Al alloy: (a), (b) optical micrographs of the transverse
central section of the rod, (c) optical micrographs of the longitudinal section of the rod, (d), (e), (f) TEM
micrographs at different magnification.
475 Analysis of structure and mechanical properties of AA 5083 aluminium alloy processed by ECAE
dislocation density distribution (Fig. 1f) resulting
from the casting process, the calculated value from
Eq. (1) is = 1.7
.
10
14
m
-2
.
3.2. Microstructure of AA5083 after
ECAP
General details of the microstructure obtained after
deformation by ECAP for 4 passes following route
C of the AA5083 aluminium alloy are presented in
Fig. 2. After this level of strain, the microstructure
shows an inhomogeneous elongated and refined
structure (Figs. 2a and 2b). Fig. 2c corresponds to
the longitudinal centred section of the deformed
sample. When comparing to Fig. 1c, grains ap-
peared more irregular and segregation seems to di-
minish or also to become more uniform. Tangled
dislocations as well as some banded zones can be
seen from TEM images of the investigated alloy
(Figs. 2d-2f). The original equiaxed as-cast grains
are deformed and have adopted a layered structure
in which some high deformed grains seem to en-
velop areas which look not as much deformed. Dif-
ferent slip system are involved during the deforma-
tion process resulting in a deformed inhomogeneous
structure and, as observed by other authors in other
Fig. 2. Microstructure of AA5083 Al alloy after 4 passes through ECAE following Route C: (a), (b) optical
micrographs of the transverse central section of the rod, (c) optical micrographs of the longitudinal section
of the rod, (d), (e), (f) TEM micrographs of the partially elongated grains at different magnification.
aluminium alloys [9,10], with two different strained
zones (Figs. 2d and 2e) as they can move as func-
tion of differences in lattice rotation between close
parts of the crystal. Also this mechanism is en-
hanced by the presence of different size second
phase particles in the AA5083 alloy as well as by
the existence of segregated areas that difficult co-
herent deformation process inside the crystal com-
plicating accommodation of the material. As ex-
pected, a notably increased in dislocation density
( is about 3.15
.
10
14
m
-2
) appears within the grains
resulting from the higher irregular strain. No lack of
cohesion at particle-matrix interface has been de-
tected. Low angle grain boundaries (LAB) are the
dominant feature encountered in the specimen.
Another set of specimens were ECAP processed
at 423K following route A until 11 passes. As can
be seen in the Fig. 3a, the processing to these high
strains leads to the formation of a complete defor-
mation structure in both transverse (Fig. 3a) and
longitudinal (Figs. 3b and 3c) sections. n addition,
second phase particles are completely broken and
oriented to the shear direction (Fig. 3d), also it is
worth noting that in some areas, high strain con-
ducted to the matrix-precipitate interface decohesion
(Fig. 3e). This fact has only been detected for the
476 N. Llorca-Isern, C. Luis-Prez, P.A. Gonzlez, L. Laborde and D. Patio
Fig. 3. Microstructure of AA5083 Al alloy after 11 passes through ECAE following Route A: (a) optical
micrographs of the highly deformed transverse central section of the rod, (b), (c) optical micrographs of the
highly deformed longitudinal section of the rod, (d), (e) SEM micrographs showing the broken precipitates
and their orientation in the shear direction (f), (g), (h) TEM micrographs of the deformed grains at different
magnification.
initial coarse (Fe,Mn)
3
SiAl
12
particles. n spite of this,
a certain homogeneity in the structure is reached
as can be observed in TEM figures. Furthermore,
the calculated dislocation density decreases to
1.35
.
10
14
m
-2
. Low angle grain boundaries (LAB) are
the dominant feature encountered for this set of
specimens.
3.3. MechanicaI behaviour of ECAP
AA5083
Tensile properties were obtained for the as-cast
AA5083 and for 8 passes processed by route A
specimens of AA5083 Al alloy. Unfortunately, each
ECAPped specimen only provided material for one
single tensile test. Taken this into account, the re-
sults of the tensile test could have given a disper-
sion that has been impossible to evaluate. Despite
this fact, the tendency is clear and it was found that
much improvement in strength was obtained for the
processed alloys. One ECAE pass (route A) almost
triplicate the strength of the alloy and the latter in-
creased with further passes. The ductility showed
by this processed alloy is significantly higher after
the second pass through the ECAE and steadily
increases with further passes.
477 Analysis of structure and mechanical properties of AA 5083 aluminium alloy processed by ECAE
f we now compare the results of both ECAE
routes (Table 1), hardness is higher for the processed
samples specially for the specimens pressed fol-
lowing route C. They showed higher macrohardness
than specimens processed 11 passes by route A.
Elastic constants obtained from nanoindentation
tests carried out on the three set of samples (as
cast, AA5083-R4, AA5083-A11) showed different
characteristics depending on the processing route,
nevertheless the Elastic modulus remains constant.
Elastic dynamic modulus was obtained by measur-
ing it with NDT Ultrasonic equipment in order to evalu-
ate this property with the lower interaction with the
TabIe 1. Hardness measurements for AA5083 Al alloy before and after ECAP processing at 423K.
VCKERS AA5083-ar AA5083-routeC AA5083- route A
4 passes 11 passes
MACROHARDNESS 85. 11 2.00 267,20 2,89 226,88 1,11
TabIe 2. Tensile tests for AA5083 Al alloy ECAP at
423K.
Number of Yield strength Ductility
passes route A (MPa) (%)
90angle die
0 (as-cast) 110 16.6
1 300
2 321 19.3
4 345 19.4
6 358
8 394 19.6
sample, see Table 2. As other authors suggested,
measurements correlate very well with E
dyn
obtained
by mechanical testing equipments [11], see Table
3. This indicates that the stiffness of the specimens
did not vary resulting from the accommodation of
different effects in the samples as is observed in
their structure. Nevertheless, at the very locally
nanoindenter measured area, total indentation
depths showed the higher resistance of the speci-
mens as well as the elastic energy involved in the
test when loading and unloading up to 1000 mN
which is more important for the ECAE processed
specially the one processed following route C. Also
the plastic energy decreases and the plastic resis-
tance is doubled comparing to the as-cast speci-
men
4. SUMMARY
ECAP is effective hi introducing a high density of
dislocations in as-cast Al 5083 alloy depending on
the route imposed for deformation. Route C seems
more effective than Route A for strengthened the
alloy. Nevertheless, if the number of passes follow-
ing Route A is high, the energy involved in the strain-
ing process can damage coarse second phase pre-
cipitates causing decohesion in the precipitate-
matrix interface. A high tensile ductility (over
TabIe 3. Elastic constants from nanoindentation tests and Elastic Dynamic Modulus (E
dyn
) from NOT
Ultrasound testing.
Wt We Wr E/(1-v
2
) H
plast
hr' E V E
dyn
(nJ) (nJ) (nJ) (GPa) (N/mm
2
) (mm) (GPa) (GPa)
AA5083 2363,00 253,50 2109,17 59,58 851,17 6,67 73.33 0.35 75.78
As-received 33.53 19.54 23.77 5.70 35.89 0.14 5.00
AAS083 1790,60 304,60 1485,60 68,44 1609,20 4,85 77.85 0.35 7S.5S
Route C 24.66 9.92 31.34 2.51 23.12 0.04 2.21
4 passes
AA5083 Route A 1784,50 266,00 1518,50 78,90 1514,00 5.00 86.31 0.34 5.71
11 passes 3.54 14.14 10.61 5.09 19.80 0.03 4.47
478 N. Llorca-Isern, C. Luis-Prez, P.A. Gonzlez, L. Laborde and D. Patio
strength through strain hardening is obtained by
pressing by ECAE at 423K. An accommodation and
rearrangement of dislocations network takes place
together with the fact that high angle boundaries
(HAB) are seldom seen, thus its behaviour may be
interpreted by considering precipitates as well as
dislocation cell boundaries but not only due to the
influence of HAB. ECAE processing also contrib-
utes not only to refine material grain size but also
significantly modifying the precipitates morphology
as well as their distribution which results in a banded
microstructure and an elongated distribution of very
fine grains.
REFERENCES
[1] R.Z. Valiev, N.A. Krasilnikov and N.K.Tsenev //
Mat.Sci. Eng. A 137 (1991) 35.
[2] V.M.Segal // Mat.Sci. Eng. A 197 (1995) 157.
[3] Y.Wang, M.Chen, F.Zhou and E.Ma // Nature
419 (2002) 912.
[4] N.A.Krasilnikov, W.Lojkowski, Z.Pakiela and
R.Z. Valiev // Mat.Sci. Eng. A 397 (2005) 330.
[5] R.X. Hecht, K. Kannan, n: Superplasticity and
Superplastic Forming, ed. by A.K.Gosh and
T.R.Bieler (MS: Warrendale, 1995) p.259.
[6] Metals Reference Book, ed. M.Baucchio
(ASM: Materials Park, 1993) p.614.
[7] P.K.Ham and N.G.Sharp // Phil.Mag. 6 (1961)
1183.
[8] P.B.Hirsh, n: Electron Microscopy of Thin
Crystals, ed. by P.B.Hirsh, KB.Nicholson,
A. Howie, D.W.Pashley and M. J.Whelan
(Butterworths: London, 1965) p.532.
[9] R.Kaibyshev, K.Shilipova, F.Musin and
Y.Motohashi // Mat. Sci. Eng. A 396 (2005)
341.
[10] M.A. Munoz-Morris, C.Garcia Oca,
G.Gonzalez Doncel and D.G.Morris // Mat.
Sci. Forum 426-432 (2003) 2643.
[11] . Fonseca, J.A. Benito, . Megia, A Roca and
J. Jorba // Rev. Met. 38 (2002) 249.